Regressions happen. But you don't want to find out from app store reviews or spicy reddit comments. With our Observe service you can see which release caused a regression and what kind of devices are experiencing it. And then you can hand your agent a prompt with the data to fix it. It's good to ship and touch grass. It's more relaxing with Observe:

A simple agentic workflow we love: Feedback comes in. An agent reproduces it. Writes the fix. Runs it on a simulator. Opens the PR. You stay focused on the product. The loop handles the busywork.

I absolutely love Apple SDKs. My first language was Objective-C, I won a scholarship to WWDC, and I started my career as an iOS dev. I firmly believe the best apps feel native to the OS, using platform primitives and human interface guidelines. But we need to stop it with the class wars. Apple's tooling has atrophied and practices have become hostile. Developers expect web-like iteration speed, and want to ship the same high-quality apps on multiple platforms with small teams. I love my toolchain. We use React Native because I feel like it allows our teams to ship the level of polish and quality I've come to expect from excellent mobile apps at a fraction of the effort. I might not have the same opinion in 6 months, and you don't have to agree, but I'm tired of the tool being the measure of the developer. Use the best tool for the job, build a great app and let your work speak for itself.

We see plenty of tweets about queue times (especially during the busiest hours). So we built a page where you can self-serve more information about the queue. Check it out:

A native module is now one Swift class with @𝙹𝚂 on the parts JavaScript should see 🎉 Expo Modules 2.0 → 𝙵𝚞𝚗𝚌𝚝𝚒𝚘𝚗 and 𝙰𝚜𝚢𝚗𝚌𝙵𝚞𝚗𝚌𝚝𝚒𝚘𝚗 collapse into one @𝙹𝚂 method → 100,000 adds: 97 ms → 19 ms on the same SDK → iOS in SDK 57 today Benchmarks and code from @tsapeta:

@kzzzf It’s easier to develop a high-performance native app in React Native. SwiftUI has limited functionality, poor layout and animation performance, and DX is mid at best. It’s been discouraging to see it barely improve since released 7 years ago!

Created a GPT-Live-1 voice app template with @expo - Full-duplex audio + background calls on iOS - Choose from all supported voices - 6 Rive animations from Vercel AI Elements - Expo SDK 57 + native Expo UI controls Source below ↓

🙌 𝝠 Expo SDK 58 beta is out. Built for iOS 27 on React Native 0.88 RC, with 𝚎𝚡𝚙𝚘-𝚊𝚙𝚙-𝚒𝚗𝚝𝚎𝚗𝚝𝚜 for Siri, @𝚎𝚡𝚙𝚘/𝚊𝚐𝚎𝚗𝚝-𝚌𝚕𝚒 for coding agents, and Expo Modules 2.0 in beta....and tons more. 𝚗𝚙𝚡 𝚎𝚡𝚙𝚘 𝚒𝚗𝚜𝚝𝚊𝚕𝚕 𝚎𝚡𝚙𝚘@𝚗𝚎𝚡𝚝 --𝚏𝚒𝚡 Tons of delicious details in the changelog:

Now you can find the flakiest flow in your test suite with one command. EAS Insights for Workflows and Maestro reads straight from the terminal. 𝚎𝚊𝚜 𝚠𝚘𝚛𝚔𝚏𝚕𝚘𝚠:𝚒𝚗𝚜𝚒𝚐𝚑𝚝𝚜:𝚖𝚊𝚎𝚜𝚝𝚛𝚘 --𝚜𝚘𝚛𝚝 𝚏𝚕𝚊𝚔𝚎-𝚛𝚊𝚝𝚎

A good agentic workflow for mobile is not “AI wrote some screens.” It should be a loop: a signal hits, an agent changes the app, verifies on a device or cloud simulator, then uses release infra to build, update, or submit, with proof attached for a human to approve. On Expo that loop runs on shared building blocks (SDK, Skills/MCP, EAS Build, Workflows, Simulators, Update, Submit, Observe). WIth Expo you get the blocks and the defaults, but we don't lock you in. Using everything together is powerful but we know everyone's environment/use case is different. If you only need our cloud simulators or our OTA Update service you can just take those off the shelf. Start here:

expo-device-hub, new from @expo, turns your browser into a local device dashboard 📱 - live stream of every iOS sim + Android emulator - tap, swipe, scroll, type in the page - boot and shut down without Xcode or Android Studio - flip dark mode, and more bun expo install expo-device-hub, then bun expo start. The link prints in your terminal :)